view i386/flacdsp_mmx.c @ 6143:75c09a7168c7 libavcodec

gcc2 doesn't have bogus mergel, patch from Sigbjrn Skjret cisc...broadpark___no
author lu_zero
date Sun, 13 Jan 2008 23:58:12 +0000
parents fb99890ee609
children 0ea2b97aa9f6
line wrap: on
line source

/*
 * MMX optimized FLAC DSP utils
 * Copyright (c) 2007 Loren Merritt
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#include "dsputil_mmx.h"

static void apply_welch_window_sse2(const int32_t *data, int len, double *w_data)
{
    double c = 2.0 / (len-1.0);
    int n2 = len>>1;
    long i = -n2*sizeof(int32_t);
    long j =  n2*sizeof(int32_t);
    asm volatile(
        "movsd   %0,     %%xmm7 \n\t"
        "movapd  %1,     %%xmm6 \n\t"
        "movapd  %2,     %%xmm5 \n\t"
        "movlhps %%xmm7, %%xmm7 \n\t"
        "subpd   %%xmm5, %%xmm7 \n\t"
        "addsd   %%xmm6, %%xmm7 \n\t"
        ::"m"(c), "m"(*ff_pd_1), "m"(*ff_pd_2)
    );
#define WELCH(MOVPD)\
    asm volatile(\
        "1:                         \n\t"\
        "movapd   %%xmm7,  %%xmm1   \n\t"\
        "mulpd    %%xmm1,  %%xmm1   \n\t"\
        "movapd   %%xmm6,  %%xmm0   \n\t"\
        "subpd    %%xmm1,  %%xmm0   \n\t"\
        "pshufd   $0x4e,   %%xmm0, %%xmm1 \n\t"\
        "cvtpi2pd (%4,%0), %%xmm2   \n\t"\
        "cvtpi2pd (%5,%1), %%xmm3   \n\t"\
        "mulpd    %%xmm0,  %%xmm2   \n\t"\
        "mulpd    %%xmm1,  %%xmm3   \n\t"\
        "movapd   %%xmm2, (%2,%0,2) \n\t"\
        MOVPD"    %%xmm3, (%3,%1,2) \n\t"\
        "subpd    %%xmm5,  %%xmm7   \n\t"\
        "sub      $8,      %1       \n\t"\
        "add      $8,      %0       \n\t"\
        "jl 1b                      \n\t"\
        :"+&r"(i), "+&r"(j)\
        :"r"(w_data+n2), "r"(w_data+len-2-n2),\
         "r"(data+n2), "r"(data+len-2-n2)\
    );
    if(len&1)
        WELCH("movupd")
    else
        WELCH("movapd")
#undef WELCH
}

void ff_flac_compute_autocorr_sse2(const int32_t *data, int len, int lag,
                                   double *autoc)
{
    double tmp[len + lag + 2];
    double *data1 = tmp + lag;
    int j;

    if((long)data1 & 15)
        data1++;

    apply_welch_window_sse2(data, len, data1);

    for(j=0; j<lag; j++)
        data1[j-lag]= 0.0;
    data1[len] = 0.0;

    for(j=0; j<lag; j+=2){
        long i = -len*sizeof(double);
        if(j == lag-2) {
            asm volatile(
                "movsd     %6,     %%xmm0 \n\t"
                "movsd     %6,     %%xmm1 \n\t"
                "movsd     %6,     %%xmm2 \n\t"
                "1:                       \n\t"
                "movapd   (%4,%0), %%xmm3 \n\t"
                "movupd -8(%5,%0), %%xmm4 \n\t"
                "movapd   (%5,%0), %%xmm5 \n\t"
                "mulpd     %%xmm3, %%xmm4 \n\t"
                "mulpd     %%xmm3, %%xmm5 \n\t"
                "mulpd -16(%5,%0), %%xmm3 \n\t"
                "addpd     %%xmm4, %%xmm1 \n\t"
                "addpd     %%xmm5, %%xmm0 \n\t"
                "addpd     %%xmm3, %%xmm2 \n\t"
                "add       $16,    %0     \n\t"
                "jl 1b                    \n\t"
                "movhlps   %%xmm0, %%xmm3 \n\t"
                "movhlps   %%xmm1, %%xmm4 \n\t"
                "movhlps   %%xmm2, %%xmm5 \n\t"
                "addsd     %%xmm3, %%xmm0 \n\t"
                "addsd     %%xmm4, %%xmm1 \n\t"
                "addsd     %%xmm5, %%xmm2 \n\t"
                "movsd     %%xmm0, %1     \n\t"
                "movsd     %%xmm1, %2     \n\t"
                "movsd     %%xmm2, %3     \n\t"
                :"+&r"(i), "=m"(autoc[j]), "=m"(autoc[j+1]), "=m"(autoc[j+2])
                :"r"(data1+len), "r"(data1+len-j), "m"(*ff_pd_1)
            );
        } else {
            asm volatile(
                "movsd     %5,     %%xmm0 \n\t"
                "movsd     %5,     %%xmm1 \n\t"
                "1:                       \n\t"
                "movapd   (%3,%0), %%xmm3 \n\t"
                "movupd -8(%4,%0), %%xmm4 \n\t"
                "mulpd     %%xmm3, %%xmm4 \n\t"
                "mulpd    (%4,%0), %%xmm3 \n\t"
                "addpd     %%xmm4, %%xmm1 \n\t"
                "addpd     %%xmm3, %%xmm0 \n\t"
                "add       $16,    %0     \n\t"
                "jl 1b                    \n\t"
                "movhlps   %%xmm0, %%xmm3 \n\t"
                "movhlps   %%xmm1, %%xmm4 \n\t"
                "addsd     %%xmm3, %%xmm0 \n\t"
                "addsd     %%xmm4, %%xmm1 \n\t"
                "movsd     %%xmm0, %1     \n\t"
                "movsd     %%xmm1, %2     \n\t"
                :"+&r"(i), "=m"(autoc[j]), "=m"(autoc[j+1])
                :"r"(data1+len), "r"(data1+len-j), "m"(*ff_pd_1)
            );
        }
    }
}